Analyse der Matoma-HaNS-Daten

Author

Sebastian Sauer

Published

June 13, 2025

1 Hintergrund

Dieser Arbeitsbericht schildert das technische Vorgehen im Rahmen der Analyse der Matomo-Daten des BMBF-Projekt “HaNS”.

1.1 Vorgehen

Die Matomo-Klickdaten aller Semester der Projektlaufzeit wurden für diese Analyse verarbeitet. Mit Hilfe einer R-Pipeline wurden eine Reihe von Forschungsfragen analysiert.

Der komplette Code ist online dokumentiert unter https://github.com/sebastiansauer/hans. Aus Datenschutzgründen sind online keine Daten eingestellt.

Die zentrale Analyse-Pipeline-Datei ist https://github.com/sebastiansauer/hans/blob/main/_targets.R.

1.2 Forschungsfragen

  1. Wie viele Nutzer gibt es und in welchem Zeitraum?
  2. In welcher Frequenz wird HaNS aufgesucht? Wie groß sind die zeitlichen Zwischenräume zwischen der Benutzung der Plattform?
  3. Wie oft wird HaNS pro Zeitraum (z.B. Monat) besucht?
  4. Wie verändert sich die Nutzung im Zeitverlauf?
  5. Wie viele Aktionen bringt ein Visit mit sich? Wie ist die statistische Verteilung der Aktionen pro Visit?
  6. Wie lang verweilen die Nutzer pro Visit?
  7. Wie verändert sich die Nutzungsdauer pro Visit im Zeitverlauf?
  8. Welche Aktionen führen die Nutzer auf Hans aus?
  9. Wie verändern sich die Verteilungen der Aktionshäufigkeiten im Zeitverlauf?
  10. An welchen Tagen und zu welcher Zeit kommen die User zu HaNS?
  11. Wie häufig und in welcher Art inteagieren die Nutzer mit dem LLM in HaNS?
  12. Wie groß ist der Anteil der Nutzer, die mit dem LLM interagieren?
  13. Wie verändert sich der Anteil der Nutzer, die mit dem LLM interagieren, im Zeitverlauf?
  14. Wie oft wird auf ein Wort im Transkript des LLM geklickt?
  15. Wie oft wird ein Transkript-Dienst in HaNS in Anspruch genommen?
  16. Wie verändert sich die Nutzung der Transkript-Dienste in HaNS im Zeitverlauf?
  17. Wie lange werden Videos angeschaut?
  18. Wie verändert sich die Betrachtungsdauer im Zeitverlauf?

2 Setup

2.1 R-Pakete starten

Show the code 
library(targets)
library(tidyverse)
library(ggokabeito)
library(easystats)
library(gt)
library(ggfittext)
library(scales)
library(visdat)
library(collapse)
library(ggpubr)
library(knitr)
Show the code 
theme_set(theme_minimal())

2.2 Optionen setzen

Show the code 
options(lubridate.week.start = 1)  # Monday as first day
#options(collapse_mask = "all") # use collapse for all dplyr operations
options(chromote.headless = "new")  # Chrome headleass needed for gtsave

2.3 Roh-Daten laden und inspizieren (data_all_fact)

Show the code 
tar_load(data_all_fct)

2.3.1 Dimension

Der Roh-Datensatz verfügt über

  • 19086 Zeilen
  • 8399 Spalten (Dubletten und Spalten mit Bildern bereits entfernt)

Jede Zeile entspricht einem “Visit”.

2.3.2 Erster Blick

Show the code 
data_all_fct_head100 <- 
data_all_fct %>% 
  select(1:100) %>% 
  slice_head(n = 100)
Show the code 
data_all_fct_head100 %>% 
  visdat::vis_dat()

2.3.3 Namen (1-100)

Show the code 
data_all_fct_head100 %>% 
  names()
  [1] "file_id"                                 
  [2] "idvisit"                                 
  [3] "visitip"                                 
  [4] "visitorid"                               
  [5] "fingerprint"                             
  [6] "actiondetails_0_type"                    
  [7] "actiondetails_0_url"                     
  [8] "actiondetails_0_pageidaction"            
  [9] "actiondetails_0_idpageview"              
 [10] "actiondetails_0_servertimepretty"        
 [11] "actiondetails_0_pageid"                  
 [12] "actiondetails_0_sitesearchkeyword"       
 [13] "actiondetails_0_sitesearchcount"         
 [14] "actiondetails_0_pageviewposition"        
 [15] "actiondetails_0_title"                   
 [16] "actiondetails_0_subtitle"                
 [17] "actiondetails_0_timestamp"               
 [18] "sitecurrency"                            
 [19] "sitecurrencysymbol"                      
 [20] "serverdate"                              
 [21] "visitserverhour"                         
 [22] "lastactiontimestamp"                     
 [23] "lastactiondatetime"                      
 [24] "servertimestamp"                         
 [25] "firstactiontimestamp"                    
 [26] "servertimepretty"                        
 [27] "serverdatepretty"                        
 [28] "serverdateprettyfirstaction"             
 [29] "servertimeprettyfirstaction"             
 [30] "visitortype"                             
 [31] "visitcount"                              
 [32] "dayssincefirstvisit"                     
 [33] "secondssincefirstvisit"                  
 [34] "visitduration"                           
 [35] "visitdurationpretty"                     
 [36] "searches"                                
 [37] "actions"                                 
 [38] "interactions"                            
 [39] "referrertype"                            
 [40] "referrertypename"                        
 [41] "referrername"                            
 [42] "referrerurl"                             
 [43] "referrersearchengineurl"                 
 [44] "languagecode"                            
 [45] "language"                                
 [46] "devicetype"                              
 [47] "devicebrand"                             
 [48] "devicemodel"                             
 [49] "operatingsystem"                         
 [50] "operatingsystemname"                     
 [51] "operatingsystemcode"                     
 [52] "operatingsystemversion"                  
 [53] "browserfamily"                           
 [54] "browserfamilydescription"                
 [55] "browser"                                 
 [56] "browsername"                             
 [57] "browsercode"                             
 [58] "browserversion"                          
 [59] "events"                                  
 [60] "continent"                               
 [61] "continentcode"                           
 [62] "country"                                 
 [63] "countrycode"                             
 [64] "countryflag"                             
 [65] "region"                                  
 [66] "regioncode"                              
 [67] "city"                                    
 [68] "location"                                
 [69] "latitude"                                
 [70] "longitude"                               
 [71] "visitlocaltime"                          
 [72] "visitlocalhour"                          
 [73] "dayssincelastvisit"                      
 [74] "secondssincelastvisit"                   
 [75] "resolution"                              
 [76] "plugins"                                 
 [77] "provider"                                
 [78] "providername"                            
 [79] "providerurl"                             
 [80] "actiondetails_0_pageloadtime"            
 [81] "actiondetails_0_timespent"               
 [82] "actiondetails_0_timespentpretty"         
 [83] "actiondetails_0_pageloadtimemilliseconds"
 [84] "actiondetails_1_type"                    
 [85] "actiondetails_1_url"                     
 [86] "actiondetails_1_pageidaction"            
 [87] "actiondetails_1_idpageview"              
 [88] "actiondetails_1_servertimepretty"        
 [89] "actiondetails_1_pageid"                  
 [90] "actiondetails_1_eventcategory"           
 [91] "actiondetails_1_eventaction"             
 [92] "actiondetails_1_pageviewposition"        
 [93] "actiondetails_1_timestamp"               
 [94] "actiondetails_1_title"                   
 [95] "actiondetails_1_subtitle"                
 [96] "actiondetails_2_type"                    
 [97] "actiondetails_2_url"                     
 [98] "actiondetails_2_pageidaction"            
 [99] "actiondetails_2_idpageview"              
[100] "actiondetails_2_servertimepretty"

2.3.4 Werte der erst 100 Spalten

Show the code 
data_all_fct_head100 %>% 
  glimpse()
Rows: 100
Columns: 100
$ file_id                                  <fct> matomo_export_2023-03-23.csv,…
$ idvisit                                  <fct> 17, 16, 15, 13, 14, 11, 10, 1…
$ visitip                                  <fct> 87.150.0.0, 217.84.0.0, 90.18…
$ visitorid                                <fct> 9b735f0eb17af6a8, c77ae8b840d…
$ fingerprint                              <fct> 86408e41c606f8f7, 0584640e01c…
$ actiondetails_0_type                     <fct> search, action, action, actio…
$ actiondetails_0_url                      <fct> NA, https://hans.th-nuernberg…
$ actiondetails_0_pageidaction             <fct> NA, 2, 2, 3, 2, 9, 3, 4, 2, 4…
$ actiondetails_0_idpageview               <fct> NA, PlZbGn, 0f7b07, A6owU2, h…
$ actiondetails_0_servertimepretty         <fct> "Mar 23, 2023 21:18:53", "Mar…
$ actiondetails_0_pageid                   <fct> 325, 324, 323, 164, 176, 145,…
$ actiondetails_0_sitesearchkeyword        <fct> "GDI", NA, NA, NA, NA, NA, NA…
$ actiondetails_0_sitesearchcount          <fct> 0, NA, NA, NA, NA, NA, NA, NA…
$ actiondetails_0_pageviewposition         <fct> 1, 1, 1, 1, 1, 1, 1, NA, 1, N…
$ actiondetails_0_title                    <fct> Site Search, HAnS, HAnS, HAnS…
$ actiondetails_0_subtitle                 <fct> "GDI", "https://hans.th-nuern…
$ actiondetails_0_timestamp                <fct> 2023-03-23 21:18:53, 2023-03-…
$ sitecurrency                             <fct> USD, USD, USD, USD, USD, USD,…
$ sitecurrencysymbol                       <fct> $, $, $, $, $, $, $, $, $, $,…
$ serverdate                               <fct> 2023-03-23, 2023-03-23, 2023-…
$ visitserverhour                          <fct> 20, 20, 18, 18, 17, 17, 16, 1…
$ lastactiontimestamp                      <fct> 2023-03-23 20:18:53, 2023-03-…
$ lastactiondatetime                       <fct> 2023-03-23 20:18:53, 2023-03-…
$ servertimestamp                          <fct> 2023-03-23 20:18:53, 2023-03-…
$ firstactiontimestamp                     <fct> 2023-03-23 20:18:53, 2023-03-…
$ servertimepretty                         <fct> 21:18:53, 21:01:30, 19:56:41,…
$ serverdatepretty                         <fct> "Thursday, March 23, 2023", "…
$ serverdateprettyfirstaction              <fct> "Thursday, March 23, 2023", "…
$ servertimeprettyfirstaction              <fct> 21:18:53, 21:01:30, 19:56:41,…
$ visitortype                              <fct> new, new, new, returning, new…
$ visitcount                               <fct> 1, 1, 1, 4, 1, 3, 1, 3, 1, 2,…
$ dayssincefirstvisit                      <fct> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
$ secondssincefirstvisit                   <fct> 0, 0, 0, 15464, 0, 10678, 0, …
$ visitduration                            <fct> 0, 20, 3, 1494, 0, 2555, 79, …
$ visitdurationpretty                      <fct> 0s, 20s, 3s, 24 min 54s, 0s, …
$ searches                                 <fct> 1, 0, 0, 7, 0, 2, 1, 0, 0, 9,…
$ actions                                  <fct> 1, 1, 1, 158, 1, 12, 9, 1, 1,…
$ interactions                             <fct> 1, 1, 1, 38, 1, 7, 5, 0, 1, 5…
$ referrertype                             <fct> website, website, website, di…
$ referrertypename                         <fct> Websites, Websites, Websites,…
$ referrername                             <fct> elearning.ohmportal.de, elear…
$ referrerurl                              <fct> https://elearning.ohmportal.d…
$ referrersearchengineurl                  <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ languagecode                             <fct> de-de, de, de-de, en-us, de-d…
$ language                                 <fct> Language code de-de, German, …
$ devicetype                               <fct> Desktop, Desktop, Desktop, De…
$ devicebrand                              <fct> Apple, Unknown, Unknown, Appl…
$ devicemodel                              <fct> Generic Desktop, Generic Desk…
$ operatingsystem                          <fct> Mac 10.15, Windows 11, Window…
$ operatingsystemname                      <fct> Mac, Windows, Windows, Mac, i…
$ operatingsystemcode                      <fct> MAC, WIN, WIN, MAC, IOS, MAC,…
$ operatingsystemversion                   <fct> 10.15, 11, 10, 10.15, 16.3, 1…
$ browserfamily                            <fct> WebKit, Blink, Blink, Gecko, …
$ browserfamilydescription                 <fct> "WebKit (Safari)", "Blink (Ch…
$ browser                                  <fct> Safari 16.3, Microsoft Edge 1…
$ browsername                              <fct> Safari, Microsoft Edge, Chrom…
$ browsercode                              <fct> SF, PS, CH, FF, MF, FF, PS, F…
$ browserversion                           <fct> 16.3, 111, 111, 111, 16.3, 11…
$ events                                   <fct> 0, 0, 0, 120, 0, 5, 4, 1, 0, …
$ continent                                <fct> Europe, Europe, Europe, North…
$ continentcode                            <fct> eur, eur, eur, amn, eur, amn,…
$ country                                  <fct> Germany, Germany, Germany, Un…
$ countrycode                              <fct> de, de, de, us, de, us, de, u…
$ countryflag                              <fct> plugins/Morpheus/icons/dist/f…
$ region                                   <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ regioncode                               <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ city                                     <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ location                                 <fct> "Germany", "Germany", "German…
$ latitude                                 <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ longitude                                <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ visitlocaltime                           <fct> 21:18:53, 21:01:29, 19:56:40,…
$ visitlocalhour                           <fct> 21, 21, 19, 18, 18, 17, 17, 1…
$ dayssincelastvisit                       <fct> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
$ secondssincelastvisit                    <fct> 0, 0, 0, 4786, 0, 7331, 0, 73…
$ resolution                               <fct> 810x1080, 1536x864, 1920x1080…
$ plugins                                  <fct> "cookie", "cookie, pdf", "coo…
$ provider                                 <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ providername                             <fct> "Unknown", "Unknown", "Unknow…
$ providerurl                              <fct> NA, NA, NA, NA, NA, NA, NA, N…
$ actiondetails_0_pageloadtime             <fct> NA, 2.23s, 0.55s, NA, 1s, NA,…
$ actiondetails_0_timespent                <fct> NA, 20, 0, 3, 0, 25, 4, NA, 0…
$ actiondetails_0_timespentpretty          <fct> NA, 20s, 0s, 3s, 0s, 25s, 4s,…
$ actiondetails_0_pageloadtimemilliseconds <fct> NA, 2233, 550, NA, 1001, NA, …
$ actiondetails_1_type                     <fct> NA, NA, NA, event, NA, event,…
$ actiondetails_1_url                      <fct> NA, NA, NA, "https://hans.th-…
$ actiondetails_1_pageidaction             <fct> NA, NA, NA, 10, NA, 10, 4, NA…
$ actiondetails_1_idpageview               <fct> NA, NA, NA, w2vQcb, NA, ZZd1q…
$ actiondetails_1_servertimepretty         <fct> NA, NA, NA, "Mar 23, 2023 18:…
$ actiondetails_1_pageid                   <fct> NA, NA, NA, 165, NA, 147, 143…
$ actiondetails_1_eventcategory            <fct> NA, NA, NA, click_button, NA,…
$ actiondetails_1_eventaction              <fct> NA, NA, NA, "Medien", NA, "ET…
$ actiondetails_1_pageviewposition         <fct> NA, NA, NA, 1, NA, 1, 1, NA, …
$ actiondetails_1_timestamp                <fct> NA, NA, NA, 2023-03-23 18:37:…
$ actiondetails_1_title                    <fct> NA, NA, NA, Event, NA, Event,…
$ actiondetails_1_subtitle                 <fct> NA, NA, NA, "Category: \"\"cl…
$ actiondetails_2_type                     <fct> NA, NA, NA, action, NA, searc…
$ actiondetails_2_url                      <fct> NA, NA, NA, "https://hans.th-…
$ actiondetails_2_pageidaction             <fct> NA, NA, NA, 9, NA, NA, 2, NA,…
$ actiondetails_2_idpageview               <fct> NA, NA, NA, EGozRT, NA, ZZd1q…
$ actiondetails_2_servertimepretty         <fct> NA, NA, NA, "Mar 23, 2023 18:…

2.3.5 Datensatz data_slim_filtered, Zeilen 1-100

Show the code 
tar_load(data_slim_filtered)

data_slim_filtered %>% 
  slice(1:100) |> 
  gt()
nr type value idvisit
0 subtitle https://hans.th-nuernberg.de/login 1
0 subtitle https://hans.th-nuernberg.de/login?evalId=none&user=undefined&role=undefined 1
0 subtitle https://hans.th-nuernberg.de/ 1
0 subtitle https://hans.th-nuernberg.de/login?evalId=none&role=undefined 1
0 timestamp 2023-03-23 14:20:12 1
0 timestamp 2024-06-06 11:19:16 1
0 timestamp 2022-12-05 15:33:45 1
0 timestamp 2023-10-04 16:19:46 1
1 eventcategory click_videocard 1
1 eventcategory login 1
1 eventaction Einführung in HAnS 1
1 eventaction success 1
1 timestamp 2023-03-23 14:20:26 1
1 timestamp 2024-06-06 11:21:07 1
1 timestamp 2022-12-05 15:33:49 1
1 timestamp 2023-10-04 16:19:54 1
1 subtitle https://hans.th-nuernberg.de/ 1
1 subtitle https://hans.th-nuernberg.de/login?evalId=none&user=undefined&role=undefined 1
1 subtitle Category: ""click_videocard', Action: ""Einführung in HAnS"" 1
1 subtitle Category: ""login', Action: ""success"" 1
2 subtitle Category: ""click_toggle', Action: ""Evaluation"" 1
2 subtitle https://hans.th-nuernberg.de/video-player?uuid=b57c1dfe-a667-48a6-b43d-dac7e600ae8c 1
2 subtitle https://hans.th-nuernberg.de/?evalId=none&role=developer 1
2 timestamp 2023-03-23 14:20:28 1
2 timestamp 2022-12-05 15:33:49 1
2 timestamp 2023-10-04 16:19:54 1
2 eventcategory click_toggle 1
2 eventaction Evaluation 1
3 eventcategory click_button 1
3 eventcategory videoplayer_click 1
3 eventcategory click_button 1
3 eventaction Kanäle 1
3 eventaction play 1
3 eventaction Kanäle 1
3 timestamp 2023-03-23 14:20:30 1
3 timestamp 2022-12-05 15:33:49 1
3 timestamp 2023-10-04 16:19:56 1
3 subtitle Category: ""click_button', Action: ""Kanäle"" 1
3 subtitle Category: ""videoplayer_click', Action: ""play"" 1
3 subtitle Category: ""click_button', Action: ""Kanäle"" 1
4 subtitle https://hans.th-nuernberg.de/channels 1
4 subtitle https://hans.th-nuernberg.de/ 1
4 subtitle https://hans.th-nuernberg.de/channels?evalId=none&role=developer 1
4 timestamp 2023-03-23 14:20:30 1
4 timestamp 2022-12-05 15:33:52 1
4 timestamp 2023-10-04 16:19:56 1
5 eventcategory click_button 1
5 eventcategory click_button 1
5 eventaction Medien 1
5 eventaction Medien 1
5 timestamp 2023-03-23 14:20:31 1
5 timestamp 2022-12-05 15:33:52 1
5 timestamp 2023-10-04 16:21:23 1
5 subtitle https://hans.th-nuernberg.de/ 1
5 subtitle Category: ""click_button', Action: ""Medien"" 1
5 subtitle Category: ""click_button', Action: ""Medien"" 1
5 pagetitle HAnS 1
6 eventcategory click_button 1
6 eventaction Medien 1
6 timestamp 2023-03-23 14:20:31 1
6 timestamp 2022-12-05 15:55:43 1
6 timestamp 2023-10-04 16:21:23 1
6 subtitle Category: ""click_button', Action: ""Medien"" 1
6 subtitle https://hans.th-nuernberg.de/ 1
6 subtitle https://hans.th-nuernberg.de/?evalId=none&role=developer 1
6 pagetitle HAnS 1
6 pagetitle HAnS 1
7 pagetitle HAnS 1
7 subtitle Category: ""click_toggle', Action: ""Evaluation"" 1
7 subtitle https://hans.th-nuernberg.de/ 1
7 subtitle Category: ""click_button', Action: ""Medien"" 1
7 timestamp 2023-03-23 14:20:34 1
7 timestamp 2022-12-05 16:03:01 1
7 timestamp 2023-10-04 16:25:22 1
7 eventcategory click_toggle 1
7 eventcategory click_button 1
7 eventaction Evaluation 1
7 eventaction Medien 1
8 subtitle https://hans.th-nuernberg.de/ 1
8 subtitle https://hans.th-nuernberg.de/ 1
8 subtitle Category: ""click_button', Action: ""Kanäle"" 1
8 timestamp 2023-03-23 14:21:47 1
8 timestamp 2022-12-05 16:03:06 1
8 timestamp 2023-10-04 16:25:23 1
8 eventcategory click_button 1
8 eventaction Kanäle 1
9 eventcategory click_toggle 1
9 eventaction Evaluation 1
9 timestamp 2023-03-23 14:22:41 1
9 timestamp 2022-12-05 16:03:16 1
9 timestamp 2023-10-04 16:25:23 1
9 subtitle Category: ""click_toggle', Action: ""Evaluation"" 1
9 subtitle https://hans.th-nuernberg.de/video-player?uuid=b57c1dfe-a667-48a6-b43d-dac7e600ae8c 1
9 subtitle https://hans.th-nuernberg.de/channels?evalId=none&role=developer 1
10 eventcategory videoplayer_click 1
10 eventaction play 1
10 timestamp 2023-03-23 14:22:45 1
10 timestamp 2022-12-05 16:03:16 1
10 timestamp 2023-10-04 16:25:24 1
10 subtitle https://hans.th-nuernberg.de/channels 1

3 Datensatz nur User

Entfernt man Developer, Admins und Lecturers aus dem Roh-Datensatz so bleiben weniger Zeilen übrig:

Show the code 
tar_load(data_users_only)
  • 11383 Zeilen
  • 8399 Spalten

4 Datensatz mit Anzahl der Aktionen pro User

Show the code 
tar_load(count_action)

5 Zeitraum

5.1 Beginn/Ende der Daten

Show the code 
tar_load(config)
Show the code 
tar_load(time_minmax)
Show the code 
time_minmax |> 
  summarise(time_min = min(time_min, na.rm = T),
            time_max = max(time_max, na.rm = T)) |> 
  gt()
time_min time_max
2022-12-05 15:33:45 2025-03-03 14:08:54

Diese Statistik wurde auf Basis des Datenobjekts data_slim_filtered berechnet.

5.2 Days since last visit

Show the code 
tar_load(time_visit_wday)
tar_load(time_since_last_visit)
Show the code 
time_since_last_visit <- 
time_since_last_visit |> 
  mutate(dayssincelastvisit = as.numeric(dayssincelastvisit)) 

time_since_last_visit |> 
  datawizard::describe_distribution(dayssincelastvisit) |> 
  knitr::kable(digits = 2)
Variable Mean SD IQR Min Max Skewness Kurtosis n n_Missing
dayssincelastvisit 6.3 15.38 0 1 92 3.32 10.84 18924 162
Show the code 
time_since_last_visit |>
  ggplot(aes(x=dayssincelastvisit)) +
  geom_density()

5.3 Visits im Zeitverlauf

Wie viele Visits (von Hans) gab es?

5.3.1 Pro Monat

Show the code 
time_visit_wday_summary <- 
time_visit_wday |> 
  ungroup() |> 
  mutate(month_start = floor_date(date_time, "month")) |> 
  mutate(month_name = month(date_time, label = TRUE, abbr = FALSE),
         month_num = month(date_time, label = FALSE),
         year_num = year(date_time)) |> 
  group_by(year_num, month_num) |> 
  summarise(n = n())
Show the code 
time_visit_wday_summary |> 
  gt()
month_num n
2022
12 25
2023
1 20
2 77
3 99
4 219
5 300
6 246
7 390
8 20
9 27
10 22
11 8
12 98
2024
1 552
2 71
3 51
4 94
5 303
6 200
7 192
8 6
9 5
10 8
11 149
12 650
2025
1 830
2 131
3 2
Show the code 
time_visit_wday_summary2 <- 
time_visit_wday |> 
  ungroup() |> 
  mutate(month_start = floor_date(date_time, "month")) |> 
  mutate(month_name = month(date_time, label = TRUE, abbr = FALSE),
         month_num = month(date_time, label = FALSE),
         year_num = year(date_time)) |> 
  group_by(year_num, month_start) |> 
  summarise(n = n())


time_visit_wday_summary2 |> 
  ggplot(aes(x = month_start, y = n)) +
  geom_line(group = 1, color = "grey60") +
  geom_point()

5.3.2 Akkumulierte Seitenaufrufe im Zeitverlauf

Show the code 
time_visit_wday_summary2 <- 
time_visit_wday |> 
  ungroup() |> 
  mutate(month_start = floor_date(date_time, "month")) |> 
  mutate(month_name = month(date_time, label = TRUE, abbr = FALSE),
         month_num = month(date_time, label = FALSE),
         year_num = year(date_time)) |> 
  group_by(year_num, month_start) |> 
  summarise(n = n()) |> 
  ungroup() |> 
  mutate(n_cumsum = cumsum(n))


time_visit_wday_summary2 |> 
  ggplot(aes(x = month_start, y = n_cumsum)) +
  geom_line(group = 1, color = "grey60") +
  geom_point()

5.4 Statistiken

Die folgenden Statistiken beruhen auf dem Datensatz data_slim_filtered:

Show the code 
glimpse(data_slim_filtered)
Rows: 3,597,119
Columns: 4
$ nr      <int> 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2,…
$ type    <fct> subtitle, subtitle, subtitle, subtitle, timestamp, timestamp, …
$ value   <chr> "https://hans.th-nuernberg.de/login", "https://hans.th-nuernbe…
$ idvisit <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…

nr fasst die Nummer der Aktion innerhalb eines bestimmten Visits.

5.5 Mit allen Daten (den 499er-Daten)

Show the code 
tbl_count_action <- 
count_action |> 
  describe_distribution(nr_max) |> 
  gt() |> 
  fmt_number(columns = where(is.numeric),
             decimals = 2)

tbl_count_action
Variable Mean SD IQR Min Max Skewness Kurtosis n n_Missing
nr_max 129.35 152.20 161.00 1.00 499.00 1.40 0.77 4,795.00 0.00
Show the code 
gtsave(tbl_count_action, filename = "reports/tbl_count_action.png")

nr_max gibt den Maximalwert von nr zurück, sagt also, wie viele Aktionen maximal von einem Visitor ausgeführt wurden.

Betrachtet man die Anzahl der Aktionen pro Visitor näher, so fällt auf, dass der Maximalwert (499) sehr häufig vorkommt:

Show the code 
count_action |> 
  count(nr_max) |> 
  ggplot(aes(x = nr_max, y = n)) +
  geom_col()

Hier noch in einer anderen Darstellung:

Show the code 
count_action |> 
  count(nr_max) |> 
  ggplot(aes(x = nr_max, y = n)) +
  geom_point()

Der Maximalwert ist einfach auffällig häufig:

Show the code 
count_action |> 
  count(nr_max == 499) |> 
  gt()
nr_max == 499 n
FALSE 4386
TRUE 409

Es erscheint plausibel, dass der Maximalwert alle “gekappten” (zensierten, abgeschnittenen) Werte fasst, also viele Werte, die eigentlich größer wären (aber dann zensiert wurden).

5.6 Nur Visitors, für die weniger als 500 Aktionen protokolliert sind

Show the code 
count_action2 <- 
count_action |> 
  filter(nr_max != 499) 

count_action2 |> 
  describe_distribution(nr_max) |> 
  gt() |> 
  fmt_number(columns = where(is.numeric),
             decimals = 2)
Variable Mean SD IQR Min Max Skewness Kurtosis n n_Missing
nr_max 94.89 106.74 122.00 1.00 496.00 1.56 2.01 4,386.00 0.00

5.7 Verteilung

5.8 Mit den 499er-Daten

Show the code 
count_action_avg = mean(count_action$nr_max)
count_action_sd = sd(count_action$nr_max)

count_action |> 
  ggplot() +
  geom_histogram(aes(x = nr_max)) +
  labs(x = "Anzahl von Aktionen pro Visit",
       y = "n",
       caption = "Der vertikale Strich zeigt den Mittelwert; der horizontale die SD") +
  theme_minimal() +
  geom_vline(xintercept = count_action_avg,
             color = palette_okabe_ito()[1]) +
  geom_segment(x = count_action_avg-count_action_sd,
               y = 0,
               xend = count_action_avg + count_action_sd,
               yend = 0,
               color = palette_okabe_ito()[2],
               size = 2) +
  annotate("label", x = count_action_avg, y = 1500, label = "MW") +
  annotate("label", x = count_action_avg + count_action_sd, y = 0, label = "SD")

Show the code 
  #geom_label(aes(x = count_action_avg), y = 1, label = "Mean")
  • Mittelwert der Aktionen pro Visit: 129.35.
  • SD der Aktionen pro Visit: 152.2.

5.9 Ohne 499er-Daten

Show the code 
count_action_avg2 = mean(count_action2$nr_max)
count_action_sd2 = sd(count_action2$nr_max)

count_action2 |> 
  ggplot() +
  geom_histogram(aes(x = nr_max)) +
  labs(x = "Anzahl von Aktionen pro Visit",
       y = "n",
       title = "Verteilung der User-Aktionen pro Visit",
       caption = "Der vertikale Strich zeigt den Mittelwert; der horizontale die SD") +
  theme_minimal() +
  geom_vline(xintercept = count_action_avg2,
             color = palette_okabe_ito()[1]) +
  geom_segment(x = count_action_avg-count_action_sd2,
               y = 0,
               xend = count_action_avg2 + count_action_sd2,
               yend = 0,
               color = palette_okabe_ito()[2],
               size = 2) +
  annotate("label", x = count_action_avg2, y = 1500, label = "MW", vjust = "top") +
  annotate("label", x = count_action_avg2 + count_action_sd2, y = 0, label = "SD", vjust = "bottom")

Show the code 
  #geom_label(aes(x = count_action_avg), y = 1, label = "Mean")
  • Mittelwert der Aktionen pro Visit: 94.89.
  • SD der Aktionen pro Visit: 106.74.

6 Verweildauer pro Visit

Die Visit-Zeit wurde auf 600 Min. trunkiert/begrenzt.

Show the code 
tar_load(time_spent)
tar_load(time_duration)

time_spent <- 
  time_spent |> 
  mutate(t_min = as.numeric(time_diff, units = "mins")) |> 
  filter(t_min < 600)

6.1 Verweildauer-Statistiken in Sekunden

Show the code 
time_spent |> 
  summarise(
    mean_time_diff = round(mean(time_diff), 2),
    sd_time_diff = sd(time_diff),
    min_time_diff = min(time_diff),
    max_time_diff = max(time_diff)
  ) |> 
  summarise(
    mean_time_diff_avg = mean(mean_time_diff),
    sd_time_diff_avg = mean(sd_time_diff, na.rm = TRUE),
    min_time_diff_avg = mean(min_time_diff),
    max_time_diff_avg = mean(max_time_diff)
  ) |> 
  gt() |> 
  fmt_number(columns = everything(),
             decimals = 2)
mean_time_diff_avg sd_time_diff_avg min_time_diff_avg max_time_diff_avg
142.52 0.00 142.52 142.52
Show the code 
tar_load(time_duration)

time_duration |> 
  summarise(duration_sec_avg = mean(visitduration_sec, na.rm = TRUE))  |> 
  mutate(duration_min_avg = duration_sec_avg / 60)
  duration_sec_avg duration_min_avg
1         1289.434         21.49056

6.2 Verweildauer-Statistiken in Minuten

Show the code 
time_spent_summary <- 
time_spent |> 
  summarise(
    mean_t_min = mean(t_min),
    sd_t_min = sd(t_min),
    min_t_min = min(t_min),
    max_t_min = max(t_min)
  ) |> 
   summarise(
    mean_t_min_avg = mean(mean_t_min),
    sd_t_min_avg = mean(sd_t_min, na.rm = TRUE),
    min_t_min_avg = mean(min_t_min),
    max_t_min_avg = mean(max_t_min)
  )

time_spent_summary |> 
  gt() |> 
  fmt_number(columns = everything(),
             decimals = 2)
mean_t_min_avg sd_t_min_avg min_t_min_avg max_t_min_avg
142.52 0.00 142.52 142.52
Show the code 
small_padding_theme <- ggpubr::ttheme(
  tbody.style = tbody_style(size = 8), # Smaller font size can help
  colnames.style = colnames_style(size = 9, face = "bold"),
  padding = unit(c(2, 2), "mm") # Reduce horizontal and vertical padding
)
Show the code 
ggpubr::ggtexttable(time_spent_summary,
                    rows = NULL,
                    theme = small_padding_theme)

6.3 Visualisierung der Verweildauer

6.3.1 bins=20

Show the code 
time_spent |> 
  ggplot(aes(x = t_min)) +
  geom_histogram() +
  scale_x_time() +
  theme_minimal() +
  labs(y = "n",
       x = "Verweildauer in HaNS pro Visit in Minuten")

6.3.2 bins=100

Show the code 
time_spent |> 
  ggplot(aes(x = t_min)) +
  geom_histogram(binwidth = 5) +
  theme_minimal() +
  labs(y = "n",
       x = "Verweildauer in Minuten",
       title = "Verweildauer in HaNS pro Visit",
       caption = "binwidth = 5 Min.")

6.3.3 Zeitdauer begrenzt auf 1-120 Min.

Show the code 
time_spent2 <- 
time_spent |> 
  filter(t_min > 1, t_min < 120) 

time_spent2 |> 
  ggplot(aes(x = t_min)) +
  geom_histogram(binwidth = 10) +
  theme_minimal() +
  labs(y = "n",
       x = "Verweildauer in HaNS pro Visit in Minuten",
       title = "Verweildauer begrenzt auf 1-120 Minuten",
       caption = "bindwidth = 10 Min.")

6.3.4 Veränderung der Verweildauer im Zeitverlauf

Show the code 
time_spent_by_month <-
  time_spent |> 
  mutate(date = ymd_hms(value)) |> 
  mutate(month_start = floor_date(date, "month")) |> 
  mutate(month_name = month(month_start, label = TRUE, abbr = FALSE),
         month_num = month(month_start, label = FALSE),
         year = year(month_start)) |> 
  group_by(month_num, year) |> 
  summarise(time_spent_month_avg = mean(t_min, na.rm = TRUE),
            time_spent_month_sd = sd(t_min, na.rm = TRUE)) |> 
  arrange(year, month_num)

time_spent_by_month |> 
  gt() |> 
  fmt_auto()
year time_spent_month_avg time_spent_month_sd
12
2,022   10       0    
2,023  268.264 196.857
2,024  251.396 176.941
2
2,023  421     180.428
2,024  309.517 202.612
2,025  214.55  172.93 
3
2,023  107.075  86.062
2,025    9       0    
4
2,023  379.868 230.16 
2,024  112       0    
7
2,023  187.708 160.938
2,024  566.692  50.806
8
2,023  129.368 105.653
2,024  582       0    
9
2,023  414.391 208.662
10
2,023  181.716 115.475
2,024   43.284  12.632
1
2,024  170.763 141.968
2,025  336.1   199.91 
5
2,024  268.909 130.843
6
2,024  336.303 182.84 
11
2,024  374.266 209.407
Show the code 
time_spent_by_month |> 
  mutate(time_spent_month_avg = round(time_spent_month_avg, 2),
         time_spent_month_sd = round(time_spent_month_sd, 2)) |> 
  ggtexttable()

Show the code 
time_spent_by_month_name <- 
time_spent |> 
  mutate(date = ymd_hms(value)) |> 
  mutate(month_start = floor_date(date, "month")) |> 
  mutate(month_name = month(month_start, label = TRUE, abbr = FALSE),
         month_num = month(month_start, label = FALSE),
         year = year(month_start)) |> 
  group_by(month_start, year) |> 
  summarise(time_spent_month_avg = mean(t_min, na.rm = TRUE),
            time_spent_month_sd = sd(t_min, na.rm = TRUE))

time_spent_by_month_name |> 
ggplot(aes(x = month_start, y = time_spent_month_avg)) +
  geom_line(group = 1, color = "grey60") +
    geom_point()

7 Was machen die User?

Show the code 
tar_load(count_action_type)

7.1 Häufigkeiten

7.1.1 Nach Kategorien

Show the code 
count_action_type_counted <- 
count_action_type |> 
  drop_na() |> 
  count(category, sort = TRUE) |> 
  mutate(prop = round(n/sum(n), 2)) 

count_action_type_counted |> 
  gt()
category n prop
video 657274 0.83
click_slideChange 61934 0.08
visit_page 40549 0.05
Media item 11046 0.01
login 4148 0.01
in_media_search 3044 0.00
Search Results Count 2668 0.00
click_topic 2623 0.00
Medien 1359 0.00
logout 1283 0.00
Kanäle 1222 0.00
GESOA 971 0.00
click_channelcard 661 0.00
Evaluation 173 0.00
Data protection 35 0.00

7.1.2 Nach Kategorien im Zeitverlauf

Show the code 
count_action_type_per_month <- 
count_action_type |> 
  select(nr, idvisit, category) |> 
  ungroup() |> 
  left_join(time_visit_wday |> ungroup()) |> 
  select(-c(dow, hour, nr)) |> 
  drop_na() |> 
  mutate(month_start = floor_date(date_time, "month")) |> 
  count(month_start, category)
Show the code 
count_action_type_per_month |> 
  gt()
month_start category n
2022-12-01 Evaluation 2
2022-12-01 GESOA 6
2022-12-01 Kanäle 11
2022-12-01 Media item 208
2022-12-01 Medien 12
2022-12-01 Search Results Count 46
2022-12-01 click_channelcard 10
2022-12-01 click_slideChange 337
2022-12-01 click_topic 13
2022-12-01 in_media_search 9
2022-12-01 login 48
2022-12-01 logout 64
2022-12-01 video 2522
2022-12-01 visit_page 423
2023-01-01 GESOA 4
2023-01-01 Kanäle 2
2023-01-01 Medien 2
2023-01-01 Search Results Count 25
2023-01-01 click_channelcard 4
2023-01-01 click_slideChange 438
2023-01-01 click_topic 19
2023-01-01 in_media_search 7
2023-01-01 login 13
2023-01-01 logout 1
2023-01-01 video 3672
2023-01-01 visit_page 179
2023-02-01 Data protection 1
2023-02-01 Kanäle 7
2023-02-01 Medien 10
2023-02-01 Search Results Count 18
2023-02-01 click_slideChange 1823
2023-02-01 click_topic 3
2023-02-01 in_media_search 73
2023-02-01 login 38
2023-02-01 logout 10
2023-02-01 video 10606
2023-02-01 visit_page 491
2023-03-01 Data protection 3
2023-03-01 Evaluation 20
2023-03-01 GESOA 57
2023-03-01 Kanäle 218
2023-03-01 Media item 655
2023-03-01 Medien 215
2023-03-01 Search Results Count 215
2023-03-01 click_channelcard 102
2023-03-01 click_slideChange 1781
2023-03-01 click_topic 109
2023-03-01 in_media_search 86
2023-03-01 login 210
2023-03-01 logout 219
2023-03-01 video 9870
2023-03-01 visit_page 2394
2023-04-01 Data protection 2
2023-04-01 Evaluation 13
2023-04-01 GESOA 70
2023-04-01 Kanäle 181
2023-04-01 Media item 2796
2023-04-01 Medien 187
2023-04-01 Search Results Count 240
2023-04-01 click_channelcard 109
2023-04-01 click_slideChange 5428
2023-04-01 click_topic 203
2023-04-01 in_media_search 100
2023-04-01 login 364
2023-04-01 logout 251
2023-04-01 video 33235
2023-04-01 visit_page 3626
2023-05-01 Evaluation 24
2023-05-01 GESOA 193
2023-05-01 Kanäle 159
2023-05-01 Media item 228
2023-05-01 Medien 132
2023-05-01 Search Results Count 372
2023-05-01 click_channelcard 62
2023-05-01 click_slideChange 8371
2023-05-01 click_topic 445
2023-05-01 in_media_search 608
2023-05-01 login 320
2023-05-01 logout 91
2023-05-01 video 64184
2023-05-01 visit_page 4564
2023-06-01 Evaluation 14
2023-06-01 GESOA 100
2023-06-01 Kanäle 95
2023-06-01 Medien 60
2023-06-01 Search Results Count 337
2023-06-01 click_channelcard 52
2023-06-01 click_slideChange 10802
2023-06-01 click_topic 280
2023-06-01 in_media_search 536
2023-06-01 login 233
2023-06-01 logout 52
2023-06-01 video 70662
2023-06-01 visit_page 3486
2023-07-01 Data protection 5
2023-07-01 Evaluation 28
2023-07-01 GESOA 79
2023-07-01 Kanäle 147
2023-07-01 Media item 68
2023-07-01 Medien 106
2023-07-01 Search Results Count 264
2023-07-01 click_channelcard 68
2023-07-01 click_slideChange 6355
2023-07-01 click_topic 162
2023-07-01 in_media_search 624
2023-07-01 login 262
2023-07-01 logout 141
2023-07-01 video 68992
2023-07-01 visit_page 3784
2023-08-01 Evaluation 1
2023-08-01 GESOA 2
2023-08-01 Kanäle 10
2023-08-01 Medien 8
2023-08-01 Search Results Count 44
2023-08-01 click_channelcard 16
2023-08-01 click_slideChange 301
2023-08-01 click_topic 4
2023-08-01 in_media_search 1
2023-08-01 login 4
2023-08-01 logout 5
2023-08-01 video 2450
2023-08-01 visit_page 199
2023-09-01 Evaluation 5
2023-09-01 GESOA 4
2023-09-01 Kanäle 22
2023-09-01 Medien 10
2023-09-01 Search Results Count 41
2023-09-01 click_channelcard 19
2023-09-01 click_slideChange 109
2023-09-01 click_topic 15
2023-09-01 in_media_search 13
2023-09-01 login 6
2023-09-01 logout 9
2023-09-01 video 3292
2023-09-01 visit_page 267
2023-10-01 Kanäle 7
2023-10-01 Media item 310
2023-10-01 Medien 16
2023-10-01 Search Results Count 5
2023-10-01 click_slideChange 159
2023-10-01 click_topic 21
2023-10-01 in_media_search 5
2023-10-01 login 39
2023-10-01 logout 44
2023-10-01 video 1233
2023-10-01 visit_page 234
2023-11-01 Kanäle 3
2023-11-01 Media item 68
2023-11-01 Medien 1
2023-11-01 Search Results Count 2
2023-11-01 click_channelcard 8
2023-11-01 click_slideChange 98
2023-11-01 click_topic 1
2023-11-01 login 7
2023-11-01 logout 2
2023-11-01 video 1172
2023-11-01 visit_page 76
2023-12-01 GESOA 4
2023-12-01 Kanäle 3
2023-12-01 Medien 34
2023-12-01 Search Results Count 9
2023-12-01 click_channelcard 3
2023-12-01 click_slideChange 884
2023-12-01 click_topic 6
2023-12-01 in_media_search 6
2023-12-01 login 32
2023-12-01 logout 2
2023-12-01 video 8081
2023-12-01 visit_page 373
2024-01-01 Evaluation 23
2024-01-01 GESOA 60
2024-01-01 Kanäle 45
2024-01-01 Media item 144
2024-01-01 Medien 72
2024-01-01 Search Results Count 156
2024-01-01 click_channelcard 14
2024-01-01 click_slideChange 7781
2024-01-01 click_topic 106
2024-01-01 in_media_search 202
2024-01-01 login 317
2024-01-01 logout 92
2024-01-01 video 51149
2024-01-01 visit_page 2830
2024-02-01 Evaluation 1
2024-02-01 GESOA 3
2024-02-01 Kanäle 3
2024-02-01 Medien 3
2024-02-01 Search Results Count 37
2024-02-01 click_channelcard 1
2024-02-01 click_slideChange 786
2024-02-01 click_topic 8
2024-02-01 in_media_search 7
2024-02-01 login 60
2024-02-01 logout 16
2024-02-01 video 12419
2024-02-01 visit_page 448
2024-03-01 Data protection 1
2024-03-01 Evaluation 4
2024-03-01 GESOA 27
2024-03-01 Kanäle 12
2024-03-01 Media item 781
2024-03-01 Medien 21
2024-03-01 Search Results Count 70
2024-03-01 click_channelcard 13
2024-03-01 click_slideChange 526
2024-03-01 click_topic 78
2024-03-01 in_media_search 8
2024-03-01 login 94
2024-03-01 logout 20
2024-03-01 video 6388
2024-03-01 visit_page 924
2024-04-01 Data protection 2
2024-04-01 Evaluation 2
2024-04-01 GESOA 10
2024-04-01 Kanäle 20
2024-04-01 Medien 31
2024-04-01 Search Results Count 84
2024-04-01 click_channelcard 12
2024-04-01 click_slideChange 1727
2024-04-01 click_topic 140
2024-04-01 in_media_search 94
2024-04-01 login 104
2024-04-01 logout 29
2024-04-01 video 17945
2024-04-01 visit_page 1012
2024-05-01 Evaluation 3
2024-05-01 GESOA 20
2024-05-01 Kanäle 31
2024-05-01 Medien 39
2024-05-01 Search Results Count 187
2024-05-01 click_channelcard 16
2024-05-01 click_slideChange 3029
2024-05-01 click_topic 138
2024-05-01 in_media_search 114
2024-05-01 login 195
2024-05-01 logout 41
2024-05-01 video 39557
2024-05-01 visit_page 1987
2024-06-01 Data protection 6
2024-06-01 Evaluation 7
2024-06-01 GESOA 39
2024-06-01 Kanäle 62
2024-06-01 Media item 818
2024-06-01 Medien 84
2024-06-01 Search Results Count 174
2024-06-01 click_channelcard 47
2024-06-01 click_slideChange 4007
2024-06-01 click_topic 200
2024-06-01 in_media_search 109
2024-06-01 login 272
2024-06-01 logout 60
2024-06-01 video 31850
2024-06-01 visit_page 2390
2024-07-01 Data protection 1
2024-07-01 Evaluation 14
2024-07-01 GESOA 114
2024-07-01 Kanäle 52
2024-07-01 Medien 51
2024-07-01 Search Results Count 216
2024-07-01 click_channelcard 25
2024-07-01 click_slideChange 4700
2024-07-01 click_topic 177
2024-07-01 in_media_search 255
2024-07-01 login 216
2024-07-01 logout 48
2024-07-01 video 48145
2024-07-01 visit_page 2320
2024-08-01 Medien 1
2024-08-01 click_slideChange 2
2024-08-01 login 2
2024-08-01 video 184
2024-08-01 visit_page 29
2024-09-01 Data protection 1
2024-09-01 click_slideChange 3
2024-09-01 in_media_search 8
2024-09-01 login 4
2024-09-01 video 454
2024-09-01 visit_page 28
2024-10-01 GESOA 6
2024-10-01 click_slideChange 107
2024-10-01 login 10
2024-10-01 logout 2
2024-10-01 video 380
2024-10-01 visit_page 36
2024-11-01 Data protection 1
2024-11-01 GESOA 28
2024-11-01 Kanäle 10
2024-11-01 Medien 10
2024-11-01 Search Results Count 3
2024-11-01 click_channelcard 3
2024-11-01 click_slideChange 2380
2024-11-01 click_topic 40
2024-11-01 in_media_search 10
2024-11-01 login 127
2024-11-01 logout 8
2024-11-01 video 9909
2024-11-01 visit_page 658
2024-12-01 Data protection 10
2024-12-01 Evaluation 7
2024-12-01 GESOA 86
2024-12-01 Kanäle 65
2024-12-01 Media item 3099
2024-12-01 Medien 92
2024-12-01 Search Results Count 60
2024-12-01 click_channelcard 43
2024-12-01 click_topic 192
2024-12-01 in_media_search 80
2024-12-01 login 495
2024-12-01 logout 28
2024-12-01 video 61655
2024-12-01 visit_page 3057
2025-01-01 Data protection 2
2025-01-01 Evaluation 3
2025-01-01 GESOA 56
2025-01-01 Kanäle 40
2025-01-01 Media item 907
2025-01-01 Medien 146
2025-01-01 Search Results Count 51
2025-01-01 click_channelcard 23
2025-01-01 click_topic 255
2025-01-01 in_media_search 69
2025-01-01 login 568
2025-01-01 logout 42
2025-01-01 video 87736
2025-01-01 visit_page 4055
2025-02-01 Evaluation 2
2025-02-01 GESOA 3
2025-02-01 Kanäle 17
2025-02-01 Media item 610
2025-02-01 Medien 13
2025-02-01 Search Results Count 12
2025-02-01 click_channelcard 11
2025-02-01 click_topic 8
2025-02-01 in_media_search 20
2025-02-01 login 106
2025-02-01 logout 2
2025-02-01 video 9526
2025-02-01 visit_page 655
2025-03-01 Media item 354
2025-03-01 Medien 3
2025-03-01 login 2
2025-03-01 logout 4
2025-03-01 video 6
2025-03-01 visit_page 24

7.1.3 Nur die Top3-Kategorien

Show the code 
tar_load(time_visit_wday)
Show the code 
count_action_type_per_month_top3 <- 
count_action_type |> 
  select(nr, idvisit, category) |> 
  ungroup() |> 
  filter(category %in% c("video", "click_slideChange", "visit_page")) |> 
  left_join(time_visit_wday |> ungroup()) |> 
  select(-c(dow, hour, nr)) |> 
  drop_na() |> 
  mutate(month_start = floor_date(date_time, "month")) |> 
  count(month_start, category)
Show the code 
count_action_type_per_month_top3 |> 
  gt()
month_start category n
2022-12-01 click_slideChange 337
2022-12-01 video 2522
2022-12-01 visit_page 423
2023-01-01 click_slideChange 438
2023-01-01 video 3672
2023-01-01 visit_page 179
2023-02-01 click_slideChange 1823
2023-02-01 video 10606
2023-02-01 visit_page 491
2023-03-01 click_slideChange 1781
2023-03-01 video 9870
2023-03-01 visit_page 2394
2023-04-01 click_slideChange 5428
2023-04-01 video 33235
2023-04-01 visit_page 3626
2023-05-01 click_slideChange 8371
2023-05-01 video 64184
2023-05-01 visit_page 4564
2023-06-01 click_slideChange 10802
2023-06-01 video 70662
2023-06-01 visit_page 3486
2023-07-01 click_slideChange 6355
2023-07-01 video 68992
2023-07-01 visit_page 3784
2023-08-01 click_slideChange 301
2023-08-01 video 2450
2023-08-01 visit_page 199
2023-09-01 click_slideChange 109
2023-09-01 video 3292
2023-09-01 visit_page 267
2023-10-01 click_slideChange 159
2023-10-01 video 1233
2023-10-01 visit_page 234
2023-11-01 click_slideChange 98
2023-11-01 video 1172
2023-11-01 visit_page 76
2023-12-01 click_slideChange 884
2023-12-01 video 8081
2023-12-01 visit_page 373
2024-01-01 click_slideChange 7781
2024-01-01 video 51149
2024-01-01 visit_page 2830
2024-02-01 click_slideChange 786
2024-02-01 video 12419
2024-02-01 visit_page 448
2024-03-01 click_slideChange 526
2024-03-01 video 6388
2024-03-01 visit_page 924
2024-04-01 click_slideChange 1727
2024-04-01 video 17945
2024-04-01 visit_page 1012
2024-05-01 click_slideChange 3029
2024-05-01 video 39557
2024-05-01 visit_page 1987
2024-06-01 click_slideChange 4007
2024-06-01 video 31850
2024-06-01 visit_page 2390
2024-07-01 click_slideChange 4700
2024-07-01 video 48145
2024-07-01 visit_page 2320
2024-08-01 click_slideChange 2
2024-08-01 video 184
2024-08-01 visit_page 29
2024-09-01 click_slideChange 3
2024-09-01 video 454
2024-09-01 visit_page 28
2024-10-01 click_slideChange 107
2024-10-01 video 380
2024-10-01 visit_page 36
2024-11-01 click_slideChange 2380
2024-11-01 video 9909
2024-11-01 visit_page 658
2024-12-01 video 61655
2024-12-01 visit_page 3057
2025-01-01 video 87736
2025-01-01 visit_page 4055
2025-02-01 video 9526
2025-02-01 visit_page 655
2025-03-01 video 6
2025-03-01 visit_page 24
Show the code 
count_action_type_per_month_top3 |> 
  ggplot(aes(x = month_start, y = n, color = category, group = category)) +
  geom_line()

7.1.4 eventcategory

Was machen die Visitors eigentlich? Und wie oft?

Show the code 
data_slim_filtered_count <- 
data_slim_filtered |> 
  filter(type == "eventcategory") |> 
  count(value, sort = TRUE) 

data_slim_filtered_count |> 
  gt()
value n
videoplayer_click 646885
clear_transcript_text_for_llm_context 104108
click_slideChange 61934
click_button 20409
click_transcript_word 7882
click_videocard 7164
verify_option_wrong 4255
login 4147
click_topic_position_using_image 2229
click_in_media_search_results 1268
in_media_search 1036
generate_questionaire 1014
generate_questionaire_interval_minutes 1014
click_start_resize 975
click_stop_resize 942
in_media_search_results 890
verify_option_correct 708
click_channelcard 587
click_videocard_search_lecturer 537
logout 519
select_transcript_text_for_llm_context 468
click_videocard_search_course_acronym 402
click_option 338
message_to_llm 279
message_to_llm_de 258
eval 240
llm_response_de 230
click_topic_position_using_link 213
press_enter 211
click_topic_details 181
click_toggle 171
click_videocard_search_semester 104
click_channelcard_search_course_acronym 80
userRole 75
click_videocard_search_course 57
click_channelcard_search_faculty_acronym 33
click_channelcard_search_lecturer 33
message_to_llm_en 27
llm_response_en 26
click_channelcard_search_course 20
click_channelcard_search_semester 10
click_channelcard_search_faculty 7
click_link 5
click_survey 3
click_chat_message_thumbs_down 1
click_chat_message_thumbs_up 1
Show the code 
data_slim_filtered_count |> 
  ggtexttable()

7.2 Verteilung

7.2.1 Rohwerte

Show the code 
count_action_type |> 
  count(category, sort = TRUE) |> 
  ggplot(aes(y = reorder(category, n), x = n)) +
  geom_col() +
  geom_bar_text() +
  labs(
    x = "User-Aktion",
    y = "Aktion",
    title = "Anzahl der User-Aktionen nach Kategorie"
  ) +
  theme_minimal() +
  scale_x_continuous(labels = scales::comma)

7.2.2 Log-Skalierung

Show the code 
count_action_type |> 
  count(category, sort = TRUE) |> 
  ggplot(aes(y = reorder(category, n), x = n)) +
  geom_col() +
  geom_bar_text() +
  labs(
    x = "Anazhl der User-Aktionen",
    y = "Aktion",
    title = "Anzahl der User-Aktionen nach Kategorie",
    caption = "Log10-Skala"
  ) +
  theme_minimal() +
  scale_x_log10()

8 An welchen Tagen und zu welcher Zeit kommen die User zu HaNS?

8.1 Setup

Show the code 
tar_load(time_visit_wday)
Show the code 
# Define a vector with the names of the days of the week
# Note: Adjust the start of the week (Sunday or Monday) as per your requirement
days_of_week <- c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday")

# Replace numbers with day names
time_visit_wday$dow2 <- factor(days_of_week[time_visit_wday$dow],
                               levels = days_of_week)

8.2 HaNS-Login nach Uhrzeit

Show the code 
time_visit_wday |> 
  as_tibble() |> 
  count(hour) |> 
  mutate(prop = n/sum(n)) |> 
  ggplot(aes(x = hour, y = prop)) +
  geom_col() +
  theme_minimal() +
  labs(
    title = "HaNS-Nutzer sind keine Frühaufsteher",
    x = "Uhrzeit",
    y = "Anteil"
  )

Show the code 
 # coord_polar()
Show the code 
time_visit_wday |> 
  as_tibble() |> 
  count(hour) |> 
  mutate(prop = n/sum(n)) |> 
  ggplot(aes(x = hour, y = prop)) +
  geom_col() +
  theme_minimal() +
  coord_polar()

8.3 Verteilung der HaNS-Besuche nach Wochentagen

Show the code 
time_visit_wday |> 
  as_tibble() |> 
  count(dow2) |> 
  mutate(prop = n/sum(n)) |> 
  ggplot(aes(x = dow2, y = prop)) +
  geom_col() +
  theme_minimal() +
  labs(title = "Verteilung der HaNS-Logins nach Wochentagen",
       x = "Wochentag",
       y = "Anteil")

Show the code 
 # coord_polar()
Show the code 
time_visit_wday |> 
  as_tibble() |> 
  count(dow2) |> 
  mutate(prop = n/sum(n)) |> 
  ggplot(aes(x = dow2, y = prop)) +
  geom_col() +
  theme_minimal() +
  labs(title = "Verteilung der HaNS-Logins nach Wochentagen",
       x = "Wochentag",
       y = "Anteil")  +
  coord_polar()

8.3.1 HaNS-Login nach Wochentagen Uhrzeit

Show the code 
time_visit_wday |> 
  as_tibble() |> 
  count(dow2, hour) |> 
  group_by(dow2) |> 
  mutate(prop = n/sum(n)) |> 
  ggplot(aes(x = hour, y = prop)) +
  geom_col() +
  facet_wrap(~ dow2) +
  theme_minimal() +
  labs(title = "Verteilung der HaNS-Logins nach Wochentagen und Uhrzeiten",
       x = "Wochentag",
       y = "Anteil")

Show the code 
 # coord_polar()
Show the code 
time_visit_wday |> 
  as_tibble() |> 
  count(dow2, hour) |> 
  group_by(dow2) |> 
  mutate(prop = n/sum(n)) |> 
  ggplot(aes(x = hour, y = prop)) +
  geom_col() +
  facet_wrap(~ dow2) +
  theme_minimal() +
  labs(title = "Verteilung der HaNS-Logins nach Wochentagen und Uhrzeiten",
       x = "Wochentag",
       y = "Anteil") +
  coord_polar()

8.4 Anzahl der Visits nach Datum (Tagen) und Uhrzeit (bin2d)

Show the code 
time2 <- 
time_visit_wday |> 
  ungroup() |> 
  mutate(date = as.Date(date_time))

time2 |> 
  ggplot(aes(x = date, y = hour)) +
  geom_bin2d(binwidth = c(1, 1)) + # (1 day, 1 hour)
  scale_x_date(date_breaks = "1 month") +
  theme(legend.position = "bottom") +
  scale_fill_viridis_c() +
  labs(caption = "Each x-bin maps to one week")

8.5 Anzahl der Visits nach Datum (Wochen) und Uhrzeit (bin2d)

Show the code 
time2 |> 
  ggplot(aes(x = date, y = hour)) +
  geom_bin2d(binwidth = c(7, 1)) +  # 1 week, 1 hour
  scale_x_date(date_breaks = "1 week", date_labels = "%W") +
  theme(legend.position = "bottom") +
  scale_fill_viridis_c()  +
  labs(x = "Week number in 2023/2024",
       caption = "Each x-bin maps to one week")

8.6 Anzahl der Visits nach Datum (Wochen) und Wochentag (bin2d)

Show the code 
time2 |> 
  ggplot(aes(x = date, y = dow)) +
  geom_bin2d(binwidth = c(7, 1)) +  # 1 week, 1 hour
  scale_x_date(date_breaks = "1 week", date_labels = "%W") +
  theme(legend.position = "bottom") +
  scale_fill_viridis_c()  +
  labs(x = "Week number in 2023/2024",
       caption = "Each x-bin maps to one week",
       y = "Day of Week") +
  scale_y_continuous(breaks = 1:7)

9 KI-Gebrauch

9.1 Interaktion mit dem LLM

9.1.1 Art und Anzahl der Interaktionen mit dem LLM

Show the code 
data_slim_filtered_ai <- 
data_slim_filtered |> 
  filter(type == "eventcategory") |> 
  filter(str_detect(value, "llm")) |> 
  count(value, sort = TRUE) |> 
  mutate(prop = n / sum(n))

data_slim_filtered_ai |> 
  gt() |> 
  fmt_auto()
value n prop
clear_transcript_text_for_llm_context 104,108  0.988
select_transcript_text_for_llm_context     468  0.004
message_to_llm     279  0.003
message_to_llm_de     258  0.002
llm_response_de     230  0.002
message_to_llm_en      27  2.562 × 10−4
llm_response_en      26  2.467 × 10−4
Show the code 
data_slim_filtered_ai |> 
  mutate(prop = round(prop, 3)) |> 
  ggtexttable()

9.1.2 Anteil Visitors, die mit dem LLM interagieren

Show the code 
data_slim_filtered_llm_interact <- 
data_slim_filtered |> 
  mutate(has_llm = str_detect(value, "llm"))  |> 
  group_by(idvisit) |> 
  summarise(llm_used_during_visit = any(has_llm == TRUE)) |> 
  count(llm_used_during_visit) |> 
  mutate(prop = round(n /sum(n), 2)) 

data_slim_filtered_llm_interact|> 
  gt()
llm_used_during_visit n prop
FALSE 4286 0.89
TRUE 509 0.11
Show the code 
data_slim_filtered_llm_interact |> 
  ggtexttable()

9.1.3 … Im Zeitverlauf

Show the code 
tar_load(idvisit_has_llm)

idvisit_has_llm_timeline <- 
idvisit_has_llm |> 
  count(year_month, uses_llm) |> 
  ungroup() |> 
  group_by(year_month) |> 
  mutate(prop = round(n/sum(n), 2)) 

idvisit_has_llm_timeline|> 
  gt()
uses_llm n prop
2022-12
FALSE 275 0.78
TRUE 79 0.22
2023-1
FALSE 406 0.76
TRUE 126 0.24
2023-10
FALSE 96 0.78
TRUE 27 0.22
2023-11
FALSE 40 0.82
TRUE 9 0.18
2023-12
FALSE 314 0.99
TRUE 3 0.01
2023-2
FALSE 554 0.84
TRUE 108 0.16
2023-3
FALSE 74 0.94
TRUE 5 0.06
2023-4
FALSE 91 0.69
TRUE 40 0.31
2023-5
FALSE 119 0.74
TRUE 41 0.26
2023-6
FALSE 67 0.83
TRUE 14 0.17
2023-7
FALSE 219 0.93
TRUE 16 0.07
2023-8
FALSE 31 1.00
2023-9
FALSE 40 1.00
2024-1
FALSE 887 0.95
TRUE 44 0.05
2024-10
FALSE 9 1.00
2024-11
FALSE 166 0.92
TRUE 15 0.08
2024-12
FALSE 794 0.92
TRUE 67 0.08
2024-2
FALSE 94 1.00
2024-3
FALSE 6 0.67
TRUE 3 0.33
2024-4
FALSE 6 0.33
TRUE 12 0.67
2024-5
FALSE 113 1.00
2024-6
FALSE 70 0.91
TRUE 7 0.09
2024-7
FALSE 23 1.00
2024-8
FALSE 1 1.00
2025-1
FALSE 1008 0.94
TRUE 60 0.06
2025-2
FALSE 166 0.96
TRUE 7 0.04
2025-3
FALSE 2 1.00
Show the code 
idvisit_has_llm_timeline |> 
  ggtexttable()

Show the code 
idvisit_has_llm |> 
  count(year_month, uses_llm) |> 
  ungroup() |> 
  group_by(year_month) |> 
  mutate(prop = n/sum(n)) |> 
  ggplot(aes(x = year_month, y = prop, color = uses_llm, groups = uses_llm)) +
  geom_point() +
  geom_line(aes(group = uses_llm)) +
  labs(title = "Visitors, die mit dem LLM interagieren im Zeitverlauf (Anteile)")

Show the code 
idvisit_has_llm |> 
  count(year_month, uses_llm) |> 
  ungroup() |> 
  group_by(year_month) |> 
  ggplot(aes(x = year_month, y = n, color = uses_llm, groups = uses_llm)) +
  geom_point() +
  geom_line(aes(group = uses_llm)) +
  labs(title = "Visitors, die mit dem LLM interagieren im Zeitverlauf (Anzahl)")

9.2 Klick auf ein Wort im Transkript

Show the code 
data_slim_filtered |> 
  filter(type == "subtitle") |> 
  filter(!is.na(value) & value != "") |> 
  count(click_transcript_word = str_detect(value, "click_transcript_word")) |> 
  mutate(prop = round(n/sum(n), 2)) |> 
  gt()
click_transcript_word n prop
FALSE 910840 0.99
TRUE 7882 0.01

9.3 KI-Aktionen

9.3.1 Insgesamt (ganzer Zeitraum)

Show the code 
tar_load(data_long)

9.3.1.1 Im Detail

Show the code 
regex_pattern <-  "Category: \"(.*?)(?=', Action)"

ai_actions_count <- 
  data_long |> 
 # slice(1:1000) |> 
  filter(str_detect(value, "transcript")) |> 
  mutate(category = str_extract(value, regex_pattern)) |> 
  select(category) |> 
  mutate(category = str_replace_all(category, "[\"']", "")) |> 
  count(category, sort = TRUE) 

ai_actions_count |> 
  gt()
category n
NA 217059
Category: clear_transcript_text_for_llm_context 104108
Category: click_transcript_word 7882
Category: select_transcript_text_for_llm_context 468
Category: click_button 20
Category: llm_response_de 3
Category: llm_response_en 2

9.3.2 KI-Klicks pro Monat

Show the code 
tar_load(ai_transcript_clicks_per_month)
Show the code 
ai_transcript_clicks_per_month_count <-
ai_transcript_clicks_per_month |> 
  count(year_month, clicks_transcript_any) |> 
  ungroup() |> 
  group_by(year_month) |> 
  mutate(prop = round(n/sum(n), 2)) 

ai_transcript_clicks_per_month_count|> 
  gt()
clicks_transcript_any n prop
2022-12
FALSE 245 0.69
TRUE 109 0.31
2023-1
FALSE 303 0.57
TRUE 229 0.43
2023-10
FALSE 98 0.80
TRUE 25 0.20
2023-11
FALSE 41 0.84
TRUE 8 0.16
2023-12
FALSE 270 0.85
TRUE 47 0.15
2023-2
FALSE 479 0.72
TRUE 183 0.28
2023-3
FALSE 58 0.73
TRUE 21 0.27
2023-4
FALSE 93 0.71
TRUE 38 0.29
2023-5
FALSE 105 0.66
TRUE 55 0.34
2023-6
FALSE 61 0.75
TRUE 20 0.25
2023-7
FALSE 150 0.64
TRUE 85 0.36
2023-8
FALSE 27 0.87
TRUE 4 0.13
2023-9
FALSE 33 0.82
TRUE 7 0.17
2024-1
FALSE 765 0.82
TRUE 166 0.18
2024-10
FALSE 9 1.00
2024-11
FALSE 163 0.90
TRUE 18 0.10
2024-12
FALSE 778 0.90
TRUE 83 0.10
2024-2
FALSE 69 0.73
TRUE 25 0.27
2024-3
FALSE 5 0.56
TRUE 4 0.44
2024-4
FALSE 16 0.89
TRUE 2 0.11
2024-5
FALSE 106 0.94
TRUE 7 0.06
2024-6
FALSE 73 0.95
TRUE 4 0.05
2024-7
FALSE 17 0.74
TRUE 6 0.26
2024-8
FALSE 1 1.00
2025-1
FALSE 982 0.92
TRUE 86 0.08
2025-2
FALSE 171 0.99
TRUE 2 0.01
2025-3
FALSE 2 1.00
Show the code 
ai_transcript_clicks_per_month_count |> 
  ggtexttable()

Show the code 
ai_transcript_clicks_per_month_count |> 
  mutate(date = ymd(paste0(year_month,"-01"))) |> 
  ggplot(aes(x = date, y = n)) +
  geom_line(group = 1) +
  geom_point()

10 Videozeit

Wie viel Zeit verbringen die Nutzer mit dem Betrachten von Videos (“Glotzdauer”)?

10.1 Glotzdauer allgemein

Achtung: Die Videozeit ist schwierig auszuwerten. Die Nutzer beenden keine Videos, in dem sie auf “Pause” drücken, sondern indem sie andere Aktionen durchführen. Dies ist aber analytisch schwer abzubilden.

Show the code 
tar_load(glotzdauer)
Show the code 
glotzdauer |> 
  # we will assume that negative glotzdauer is the as positive glotzdauer:
  mutate(time_diff = abs(time_diff)) |> 
  # without glotzdauer smaller than 10 minutes:
  filter(time_diff < 60*10) |> 
  ggplot(aes(x = time_diff)) +
  geom_histogram() +
  scale_x_time() +
  labs(x = "Time interval [minutes]",
       caption = "Only time intervals less than 10 minutes")

Show the code 
glotzdauer_prepped <- 
glotzdauer |> 
  # we will assume that negative glotzdauer is the as positive glotzdauer:
  mutate(time_diff_abs_sec = abs(as.numeric(time_diff, units = "secs"))) |> 
  # without glotzdauer smaller than 10 minutes:
  filter(time_diff_abs_sec < 60*10) |> 
  mutate(time_diff_abs_min = time_diff_abs_sec / 60) 

glotzdauer_tbl <- 
  glotzdauer_prepped |> 
  select(time_diff_abs_sec, time_diff_abs_min) |> 
  describe_distribution()

glotzdauer_tbl |> 
  kable()
Variable Mean SD IQR Min Max Skewness Kurtosis n n_Missing
time_diff_abs_sec 57.8050366 103.166468 55.0000000 0 597.00 2.74483 7.78949 3693 0
time_diff_abs_min 0.9634173 1.719441 0.9166667 0 9.95 2.74483 7.78949 3693 0
Show the code 
glotzdauer_tbl |>
  mutate(across(where(is.numeric), ~ round(., 2))) |>
  ggpubr::ggtexttable()

10.2 Glotzdauer im Zeitverlauf

Show the code 
glotzdauer_prepped_tbl <-
glotzdauer_prepped |> 
  mutate(first_of_month = floor_date(date, unit = "month")) |> 
  group_by(first_of_month) |> 
  summarise(time_diff_mean = mean(time_diff, na.rm = TRUE))
  

glotzdauer_prepped_tbl |> 
  kable()
first_of_month time_diff_mean
2022-12-01 22.736364 secs
2023-01-01 28.109244 secs
2023-02-01 15.815451 secs
2023-03-01 47.571429 secs
2023-04-01 44.633588 secs
2023-05-01 46.980582 secs
2023-06-01 52.568750 secs
2023-07-01 35.906250 secs
2023-08-01 109.500000 secs
2023-09-01 23.733333 secs
2023-10-01 23.500000 secs
2023-11-01 27.833333 secs
2023-12-01 8.804196 secs
2024-01-01 46.468619 secs
2024-02-01 18.714286 secs
2024-03-01 16.774193 secs
2024-04-01 30.059701 secs
2024-05-01 43.555556 secs
2024-06-01 53.278846 secs
2024-07-01 55.058824 secs
2024-08-01 -24.000000 secs
2024-09-01 -9.000000 secs
2024-10-01 36.750000 secs
2024-11-01 47.181818 secs
2024-12-01 60.773707 secs
2025-01-01 52.115732 secs
2025-02-01 14.147059 secs
2025-03-01 5.000000 secs
Show the code 
glotzdauer_prepped_tbl |> 
  ggplot(aes(x = first_of_month, y = time_diff_mean)) +
  geom_line()